You are on page 1of 9

infection control and hospital epidemiology

january 2009, vol. 30, no. 1

original article

Outbreak of Multidrug-Resistant Pseudomonas aeruginosa Colonization and Infection Secondary to Imperfect Intensive Care Unit Room Design
Susy Hota, MD; Zahir Hirji, MHSc; Karen Stockton, MHSc; Camille Lemieux, MD, LLB; Helen Dedier, MLT; Gideon Wolfaardt, PhD; Michael A. Gardam, MD, MSc

background. Pseudomonas aeruginosa has been increasingly recognized for its ability to cause signicant hospital-associated outbreaks, particularly since the emergence of multidrug-resistant strains. Biolm formation allows the pathogen to persist in environmental reservoirs. Thus, multiple hospital room design elements, including sink placement and design, can impact nosocomial transmission of P. aeruginosa and other pathogens. methods. From December 2004 through March 2006, 36 patients exposed to the intensive care unit or transplant units of a tertiary care hospital were infected with a multidrug-resistant strain of P. aeruginosa. All phenotypically similar isolates were examined for genetic relatedness by means of pulsed-eld gel electrophoresis. Clinical characteristics of the affected patients were collected, and a detailed epidemiological and environmental investigation of potential sources was carried out. results. Seventeen of the infected patients died within 3 months; for 12 (71%) of these patients, infection with the outbreak organism contributed to or directly caused death. The source of the outbreak was traced to hand hygiene sink drains, where biolms containing viable organisms were found. Testing by use of a commercial uorescent marker demonstrated that when the sink was used for handwashing, drain contents splashed at least 1 meter from the sink. Various attempts were made to disinfect the drains, but it was only when the sinks were renovated to prevent splashing onto surrounding areas that the outbreak was terminated. conclusion. This report highlights the importance of biolms and of sink and patient room design in the propagation of an outbreak and suggests some strategies to reduce the risks associated with hospital sinks. Infect Control Hosp Epidemiol 2009; 30:25-33

An important goal in hospital design is to create a safe environment for the delivery of patient care. While much attention is typically directed toward the potential dispersion of environmental organisms, such as Aspergillus and Legionella species, during construction activities, other pathogens, such as Pseudomonas and Serratia species, may also pose an ongoing hazard once the facility is operational. By advising on aspects of room design, patient placement, and plumbing facilities, infection control consultation can ensure that the risks of hospital-acquired infections are minimized. Pseudomonas aeruginosa has been increasingly recognized for its ability to cause signicant hospital-associated outbreaks of infection, particularly since the emergence of multidrugresistant strains. Outbreaks of multidrug-resistant P. aeruginosa colonization or infection have been reported on urology wards, a burn unit, hematology/oncology units, and adult and neonatal critical care units.1-8 Various medical devices

and environmental reservoirs have been implicated in these outbreaks, including antiseptic solutions and lotions; endoscopy equipment; ventilator apparatus; and mouth swabs.9-13 These sources can easily be eliminated once identied. A greater challenge exists if the source of an outbreak involves permanent components of the hospital physical plant, such as plumbing xtures. We describe an outbreak of multidrug-resistant P. aeruginosa infection that resulted from colonization of hand hygiene sink drains in a recently constructed tertiary care medical/ surgical intensive care unit (MSICU), transplant stepdown unit, and transplant ward. This report highlights the key role of sink design and inpatient room design in causing such an outbreak, and it outlines effective strategies to manage outbreaks of this nature. We also emphasize the challenges that surround early outbreak recognition in a complex medical care facility.

From the Department of Infection Prevention and Control, University Health Network (S.H., Z.H., K.S., C.L., H.D., M.A.G.), and the Department of Medicine, University of Toronto (S.H., M.A.G.), and the Department of Chemistry and Biology, Ryerson University (G.W.), Toronto, Ontario, Canada. Received April 10, 2008; accepted August 6, 2008; electronically published December 1, 2008. 2008 by The Society for Healthcare Epidemiology of America. All rights reserved. 0899-823X/2009/3001-0006$15.00. DOI: 10.1086/592700

26

infection control and hospital epidemiology

january 2009, vol. 30, no. 1

methods
Outbreak Overview The outbreak occurred from December 2004 through March 2006 and involved 3 hospital areas: the MSICU, the solid organ transplant stepdown unit, and the solid organ transplant ward. The outbreak strain of P. aeruginosa was phenotypically determined to be resistant to all antipseudomonal antibiotics (ie, ceftazidime, imipenem, ciprooxacin, piperacillin-tazobactam, and gentamicin), except for variable sensitivity to amikacin. Epidemiologic Investigation Possible outbreak-affected patients were dened as patients admitted to the affected area during the outbreak period who were colonized or infected with Pseudomonas isolates that matched the outbreak phenotype. An epidemiologic investigation was carried out to search for potential case-case links or casecommon environmental source links. Demographic and clinical information for affected patients was collected by means of retrospective review of electronic medical records and the laboratory information system. Environmental screening for the outbreak strain was performed on 8 occasions in the outbreak areas. Microbiologic Evaluation of Clinical and Environmental Isolates All specimens collected for culture and sensitivity testing were tested for antimicrobial susceptibilities, according to standard protocols.14 The routine panel of antimicrobials tested by means of the VITEK automated instrument (bioMe rieux) included ceftazidime, ciprooxacin, gentamicin, tobramycin, piperacillin-tazobactam, amikacin, and imipenem. Additional testing of susceptibility to meropenem and colistin was performed using the E-test (AB Biodisk). No standard guidelines for interpretation of E-test susceptibility of P. aeruginosa to colistin are available; an isolate with an E-test minimum inhibitory concentration of no more than 4 ug/mL was considered susceptible, in keeping with published recommendations.15 Molecular Characterization The initial determination of whether an isolate belonged to the outbreak strain was made on the basis of the antimicrobial resistance phenotype; pulsed-eld gel electrophoresis (PFGE) was then used to determine genetic relatedness. Isolates were digested using the restriction enzyme SpeI, with a protocol run time of 20 hours and switch times of 5.35 seconds (BioRad CHEF Mapper; Bio-Rad). Bionumerics software (Applied Maths) was used to determine phylogenetic relatedness, as the criteria previously developed by Tenover et al.16 were found to be too discriminatory.

Biolm Testing Drain plugs from 3 sink traps were carefully rinsed with sterile water to remove nonattached cells, stained with the Live/Dead BacLight kit (Molecular Probes), and examined in a fully hydrated state using confocal laser scanning microscopy. Whole sink traps were also removed, sealed with trapped water left inside, and stored at room temperature for up to 6 weeks. Replicate sections of the traps were cut off at intervals of 2 weeks; at each interval, one-half of the samples were directly stained with BacLight and examined with confocal laser scanning microscopy, and the other half were ooded with a tryptic soy broth (diluted to a concentration of 1%) and incubated for 24 hours at room temperature to determine the persistence and viability of biolm microcolonies. Hand Hygiene Sink Evaluation To determine whether sink drain contents were being dispersed onto surfaces outside of the drain itself, an emulsication of a commercially available uorescent marker (Glow Germ) was injected deep into the drain cover of a MSICU hand hygiene sink. The faucet was then turned on and the water run for 15 seconds, while handwashing took place. With light eliminated from the room and all surfaces of the room covered with black paper, the area surrounding the sink was examined for evidence of uorescent residue by use of a longwave ultraviolet light source. In order to ensure that nonspecic uorescence was absent, a pretest using the same protocol but without the uorescent marker was performed. Outbreak Control Measures The following outbreak control measures were attempted: the use of contact precautions (wearing of gowns and gloves by healthcare workers and single room isolation of the patient) for all colonized or infected cases; staff education; enhanced environmental cleaning; disinfection of hand hygiene sink drains; closure of hand hygiene sinks; and renovation of hand hygiene sinks to prevent splashing of drain contents. Setting The Toronto General Hospital is part of the University Health Network and is located in Toronto, Ontario, Canada. It is a 400-bed tertiary care center and solid organ transplantation referral center for central and eastern Canada, performing over 400 transplants per year (lung, renal, liver, heart, and multivisceral). The MSICU consists of 22 single rooms and 1 semiprivate room; the percentage of beds occupied by transplant recipients is approximately 40%. Adjacent to the MSICU is a transplant stepdown unit, with 8 single beds, for patients who are transitioning from the MSICU to the transplant ward or vice versa. The transplant ward is located 3 oors directly below the MSICU and contains 39 beds (in 17 single and 11 semiprivate rooms). A typical MSICU room layout is shown in Figure 1A. The

outbreak of p. aeruginosa infection linked to icu design

27

dedicated hand hygiene sink is approximately 1.3 meters from the head of the bed, and it is directly adjacent to the medication and sterile dressing preparatory area (Figure 1B). The sink is a wall-mounted, hands-free model with a shallow stainless steel bowl. The water spout was designed to ow water directly into the sink drain, without hitting the sides of the bowl (Figure 1C).

results
Patient Characteristics The epidemiologic curve for the preoutbreak, outbreak, and postoutbreak periods is shown in Figure 2. From December 2004 to July 2006, there were 36 patients identied as infected or colonized with the outbreak strain of P. aeruginosa. Table

1 outlines the characteristics of affected patients and types of infections. Two-thirds of affected patients (24 of 36) were considered infected with the outbreak strain, and 17 (47.2%) of the total cohort died. An independent chart review of all deaths in infected patients revealed that infection with the outbreak strain caused death in 5 (29.4%) and contributed to death in 7 (41.2%). Twenty-one (58.3%) of affected patients were identied while in the MSICU, 5 (13.9%) in the transplant stepdown unit, 4 (11.1%) on the transplant ward, and 6 (16.7%) elsewhere in the hospital building. Of note, all of the patients identied elsewhere in the building had prior exposure to the MSICU, transplant stepdown unit, or transplant ward within the outbreak period. Thirty-four (94.4%) of affected patients

gure 1. Three images from the medical surgical intensive care unit. Panel A, typical room layout; panel B, counter used for sterile procedures and medication preparation, in relation to sink; panel C, close-up of hand hygiene sink; note shallow bowl and gooseneck faucet.

28

infection control and hospital epidemiology

january 2009, vol. 30, no. 1

were considered immunocompromised (by receipt of transplant, by malignancy, or by other cause). Environmental Testing A total of 288 environmental specimens were collected and analyzed for the presence of the outbreak strain; 28 specimens yielded positive results (Table 2). No multiuse equipment was found to be contaminated. Of all sources of environmental specimens, hand hygiene sink drains accounted for the highest proportions of positive culture results (11 of 65 sink drain specimens from the MSICU were positive for the outbreak strain, as were 10 of 59 sink drain specimens from the transplant stepdown unit and 5 of 89 sink drain specimens from the transplant ward). Many drains were found to be intermittently colonized. None of 39 specimens of source water tested yielded growth of Pseudomonas species. Two external plumbing xtures were found to be positive (1 showerhead in the MSICU and 1 spout in the transplant stepdown unit). Molecular Characterization PFGE banding patterns for clinical and environmental isolates are presented in Figure 3. Two types, designated 1 and 16, were deemed sufciently genetically related to be considered involved in the outbreak. The majority of clinical isolates were of PFGE type 1.

Biolm Testing Confocal laser scanning microscopy conrmed the presence of conuent biolms, some areas more than 100 mm thick, in the samples analyzed (Figure 4). Viable multidrug-resistant P. aeruginosa isolates phenotypically consistent with the outbreak strain were recovered from these samples. Starvation experiments, in which the intact biolms were kept for up to 6 weeks before addition of a dilute nutrient solution, showed that the biolms rapidly responded: the relative abundance of viable cells in the biolms more than doubled within 24 hours after nutrient addition. Hand Hygiene Sink Evaluation Tests using the uorescent marker revealed that splashes originating from the drains of hand hygiene sinks were visible under uorescence at least 1 m from the sink. We assume that microparticles not visible through uorescence traveled further than 1 m. Most of the surfaces of adjacent medication and sterile dressing preparation areas were within the 1-m range. Sink and Room Design Interventions Disinfection of the hand hygiene sinks that yielded the outbreak strain was attempted on 2 separate occasions, as follows (Figure 2): a 7% accelerated hydrogen peroxide gel was

gure 2. Epidemiologic curve showing the rate of colonization or infection with any strain of Pseudomonas aeruginosa and with the multidrug-resistant outbreak strain in the medical/surgical intensive care unit (MSICU) and transplant units, in relation to various sink and room design interventions. 1, sinks disinfected on 2 occasions, and sinks closed in MSICU and stepdown unit; 2, sinks opened in MSICU and transplant stepdown unit, and sinks in the 3 outbreak units closed and cleaned; 3, all sinks in outbreak units closed; 4, sinks renovated; 5, sinks opened in MSICU and transplant stepdown unit; 6, sinks opened in transplant ward.

outbreak of p. aeruginosa infection linked to icu design

29

table 1. Characteristics of Cases in Outbreak of Multidrug-Resistant Pseudomonas aeruginosa Colonization and Infection Affecting 36 Patients Variable Patient characteristic Age, years Mean SD Mean (range) Male sex Location of patient at recovery of rst positive specimen MSICU Transplant stepdown unit Transplant ward Other Type of specimen yielding rst positive result Sputum Urine Blood Othera Effect of pathogen on patient Colonization Infection Immunocompromised status Yes Due to solid organ transplant Due to other immunosuppression Due to cancer No Type of solid organ transplant received, no. (% of all transplants) Liver Kidney Lung Heart Multivisceralb Underwent surgery prior to recovery of rst positive specimen Underwent invasive procedure prior to recovery of rst positive specimenc Death within 3 months of rst positive specimen Relation of multidrug-resistant P. aeruginosa infection to death Infection directly caused death Infection contributed to death Infection was unrelated to death Duration of exposure to outbreak unit for all case patients, mean SD, days Isolate characteristic Resistant to amikacin Susceptible to amikacin Value

53.4 14.3 53.4 (1980) 19 (55.6) 21 5 4 6 17 7 5 7 (58.3) (13.9) (11.1) (16.7) (47.2) (19.4) (13.9) (19.4)

12 (33.3) 24 (66.7) 21 9 4 2 8 3 6 2 2 17 (58.3) (25.0) (11.1) (5.6) (38.1) (14.3) (28.6) (9.5) (9.5) (47.2)

14 (38.9) 17 (47.2) 5 (29.4) 7 (41.2) 5 (29.4) 34.2 23.8 19 (52.8) 17 (47.2)

note. Data are no. (%) of patients, unless otherwise indicated. MSICU, medical surgical intensive care unit; SD, standard deviation. a Percutaneous transhepatic cholangiography drain, wound, or catheter tip. b One patient received a liver and lung transplant; 1 received a liver and short bowel transplant. c Cystoscopy, endoscopy, bronchoscopy, tracheostomy, or chest tube insertion.

poured into sink drains and left for 5 minutes; sink surfaces, including the interior of faucet spouts, were exposed to a 1 : 16 dilution of the same product for 5 minutes. We submerged gooseneck faucets, drain strainers, and tap covers in 250 cc accelerated hydrogen peroxide 7% solution (diluted 1 : 16) for 5 minutes; wiped bowls with accelerated hydrogen peroxide 0.05% wipes; and closed MSICU and transplant

stepdown unit patient room sinks. While sinks were closed, 2 additional attempts were made at cleaning drains of sinks with accelerated hydrogen peroxide 7% solution (diluted 1 : 16) gel product. Although postintervention cultures were sterile, several hand hygiene sinks became recolonized over time, and disinfection had no lasting impact on eradication of the outbreak strain. On the other hand, a decision to close

30

infection control and hospital epidemiology

january 2009, vol. 30, no. 1

table 2. Proportion (%) of Environmental Specimens Found Positive for the Outbreak Strain of Pseudomonas aeruginosa, by Hospital Unit Outbreak units Source of specimen Sink taps and shower heads Sink drains Equipmentb Source water note.
a

MSICU 1/27 (3.7) 11/65 (16.9) 0/16 (0.0)

Stepdown unit 1/16 (6.3) 10/59 (16.9) 0/4 (0.0)

Transplant unit 0/10 (0.0) 5/89 (5.6) 0/2 (0.0)

All 3 2/53 26/213 0/22 0/19 (3.8) (12.2) (0.0) (0.0)

Other unita 0/5 (0.0) 0/20 (0.0)

MSICU, medical/surgical intensive care unit. Cardiovascular intensive care unit or medical, surgical, or nephrology wards. b Respiratory equipment, crash cart components, intravenous monitors, patient-lifting equipment, Pyxis medicationdispensing machine, multiple use uid dispensers, ice machine, ultrasound gel, scissor hooks, and temperature probe.

all hand hygiene sinks in the outbreak areas corresponded with an immediate halt to identication of new cases. While closed, the sinks were renovated, as follows: traps were replaced; new faucet spouts were installed that did not ow directly into the drain, thereby minimizing splashback; water ow pressure was decreased; a barrier was installed between the sinks and adjacent preparatory areas (Figure 5); and patient care materials were moved move than 1 m from sinks. During this period, portable hand hygiene sinks and alcohol-

based hand gel were used. After the sink modications were complete, the uorescent marker test for splash of drain contents was repeated on 1 intensive care unit sink; it revealed no splash onto the adjacent counter or patient bed.

discussion
We report a large outbreak of colonization and infection with multidrug-resistant P. aeruginosa that resulted in signicant

gure 3. Banding patterns determined by pulsed-eld gel electrophoresis (PFGE) and a dendrogram showing the genetic relatedness of isolates of multidrug-resistant Pseudomonas aeruginosa recovered from different patients and environmental sites. Panel A, clinical isolates; panel B, environmental isolates. BAL, bronchoalveolar lavage.

outbreak of p. aeruginosa infection linked to icu design

31

utensils or hands. In our MSICU, the spout was xed to ow water directly into the sink drain. In combination with high water pressure and a very shallow sink bowl, this created a means by which Pseudomonas biolms within the drains could be disrupted, thereby transferring the viable organism to surrounding surfaces or, potentially, to the hands of healthcare workers. Others have reported taps and drains as sources of outbreaks of P. aeruginosa colonization and infection.1-4,6,7,19-25 These reports have been based on the sequential isolation of phenotypically or genotypically related strains from both sinks and clinical specimens, as in the present study.19-22 P. aeruginosa was generally impossible to eradicate using disinfection techniques alone, and replacement of sinks or sink and/or plumbing components was emphasized as a means to eliminate the organism.1-4,21,23 One group of investigators was able to successfully control an outbreak of infection with

gure 4. Confocal laser scanning micrograph showing biolms containing microcolonies of the outbreak strain of Pseudomonas aeruginosa. Distance of optical thin section from attachment surface, 30 mm; original magnication, #750.

morbidity and mortality in an immunocompromised population: 36 patients acquired the organism during an 18month period, two-thirds of whom developed invasive infections. Of the 36 affected patients, 17 (47%) died, and 12 (71%) of these deaths were directly related either to the infection or to a subsequent complication. This high attributable mortality may be a reection of the immunocompromised nature of the patient population, as well as the multidrug resistance and possible enhanced virulence of the outbreak strain. This outbreak originated in hand hygiene sink drains. In conformity with current American Institute of Architects guidelines,17 the MSICU was designed to have 1 hand hygiene sink installed per room; however, the outbreak investigation demonstrated that both the sink design and location were less than ideal. Sinks were situated sufciently close to an area where sterile procedures and medication preparation were performed to presumably allow contamination of that area through the splash of drain contents. This risk was signicantly reduced through the installation of splash barriers. The close proximity of the sink to the patient bed, while appropriate for point-of-care hand hygiene, likely enabled direct patient contamination. We assume that smaller, less visible particles traveled far further than the 1 m we were able to visualize using uorescence. We identied several issues with sink design on the affected units. Gooseneck faucets are a popular choice for hospital hand hygiene sinks because the faucet spout, when positioned the standard 10 inches (25.4 cm) above the bowl,18 is high enough to minimize inadvertent touching of the bowl by

gure 5. Sink and counter design in the medical/surgical intensive care unit where the outbreak occurred. Panel A, before renovation; panel B, after renovation.

32

infection control and hospital epidemiology

january 2009, vol. 30, no. 1

multidrug-resistant P. aeruginosa by implementing pasteurization of water taps rather than the replacement of components5; however, tap colonization was only a late source of the organism in the outbreak (other sources were found earlier on environmental screening) and, thus, presumably had less opportunity to disseminate extensively. Also, environmental and clinical surveillance ended less than 2 months after the tap contamination was identied, so no long-term follow-up information was available. In the present study, we visually demonstrated the probable mechanism of transfer of the outbreak organism to patients by means of the uorescent marker testing. We also aborted the outbreak through simple sink and room design modications to prevent splashing, without actually eradicating the organism or moving the sinks. We based this approach on the concept that biolms are resistant to traditional disinfectant methods26,27 and may, in fact, be more widespread than can be documented through visualization. The drains in our outbreak areas were proven to contain biolms that tested positive for P. aeruginosa. These biolms typically consist of a variety of microbial species that may protect pathogens from antimicrobials. Furthermore, our results showed the resilience and survival potential of biolms under prolonged conditions of no water ow, which strongly suggest that biolms can play an important role in recontamination or seeding. Replacing sinks and exposed piping would not eradicate biolm that is more distal within the plumbing system; presumably this biolm would simply recolonize new plumbing over time. Identication of the outbreak was challenging on several levels. The background prevalence of patients colonized or infected with other multidrug-resistant P. aeruginosa strains made the cluster of outbreak cases less apparent. The continuous ow of patients between the 3 affected units, and the fact that many of these patients were close contacts of one another, made it challenging to determine the mechanism of acquisition of the outbreak strain. A further delay resulted from difculty in determining the relatedness of strains. Although the antibiogram pattern of the outbreak strain was relatively unique, it would occasionally change; similar antibiogram unpredictability of related strains of P. aeruginosa has been previously reported.6,28 This made it challenging to identify possible cases requiring further investigation and PFGE typing; therefore, caution should be exercised in the use of phenotypic measures to determine relatedness. In addition, PFGE patterns frequently changed over time for both patient and environmental isolates; P. aeruginosa is known to mutate frequently, and isolates that would traditionally be considered genetically unrelated16 may actually be from the same original clone.28 Indeed, it was only after performing PFGE on many clinical and environmental isolates that we were able to identify 2 related families (types 1 and 16) of organisms that were in keeping with the clinical epidemiology of the outbreak. Our renovations were successful in preventing the reemergence of infections with the outbreak strain. Follow-up

environmental screening more than 1 year after the termination of the outbreak has shown that the organism persists in many drains in the outbreak area (data not shown); however, only 1 new infection has been identied on the previous outbreak units, in a patient at high risk, with large open wounds requiring extensive dressing changes. In conclusion, our experience has demonstrated that, in addition to ensuring adequate numbers of hand hygiene sinks, sink placement and sink design are crucially important elements in the design of hospital rooms. This point becomes especially important in critical care areas, such as intensive care units.

acknowledgments
Financial support. Funding for this outbreak investigation and management was provided by internal sources (University Health Network). Potential conicts of interest. All authors report no conicts of interest relevant to this article. Address reprint requests to Susy Hota, MD, Infection Prevention and Control, Toronto General Hospital, 9th oor, New Clinical Services Building, 200 Elizabeth Street, Toronto, Ontario M5G 2C4 (susy.hota@uhn.on.ca).

references
1. Gillespie TA, Johnson PRE, Notman AW, Coia JE, Hanson MF. Eradication of resistant Pseudomonas aeruginosa strain after a cluster of infections in a hematology/oncology unit. Clin Microbiol Infect 2000; 6: 125-130. 2. Pena C, Dominguez MA, Pujol M, Verdaguer R, Gudiol F, Ariza J. An outbreak of carbapenem-resistant Pseudomonas aeruginosa in a urology ward. Clin Microbiol Infect 2003; 9:938-943. 3. Bert F, Maubec E, Bruneau B, Berry P, Lambert-Zechovsky N. Multiresistant Pseudomonas aeruginosa outbreak associated with contaminated tap water in a neurosurgery intensive care unit. J Hosp Infect 1998; 39: 53-62. 4. Ferroni A, Nguyen L, Pron B, Quesne G, Brusset MC, Berche P. Outbreak of nosocomial urinary tract infections due to Pseudomonas aeruginosa in a pediatric surgical unit associated with tap-water contamination. J Hosp Infect 1998; 39:301-307. 5. Bukholm G, Tannaes T, Kjelsberg ABB, Smith-Erichsen N. An outbreak of multidrug-resistant Pseudomonas aeruginosa associated with increased risk of patient death in an intensive care unit. Infect Control Hosp Epidemiol 2002; 23:441-446. 6. Aumeran C, Paillard C, Robin F, et al. Pseudomonas aeruginosa and Pseudomonas putida outbreak associated with contaminated water outlets in an oncohaematology pediatric unit. J Hosp Infect 2007; 65:47-53. 7. Vianelli N, Giannini MB, Quarti C, et al. Resolution of a Pseudomonas aeruginosa outbreak in a hematology unit with the use of disposable sterile water lters. Haematologica 2006; 91:983-985. 8. Douglas MW, Mulholland K, Denyer V, Gottlieb T. Multi-drug resistant Pseudomonas aeruginosa outbreak in a burns unit an infection control study. Burns 2001; 27:131-135. 9. Silva CV, Magalhaes VD, Pereira CR, Kawagoe JY, Ikura C, Ganc AJ. Pseudo-outbreak of Pseudomonas aeruginosa and Serratia marcescens related to bronchoscopes. Infect Control Hosp Epidemiol 2003; 24:195-197. 10. Engelhart S, Krizek L, Glasmacher A, Fischnaller E, Marklein G, Exner M. Pseudomonas aeruginosa outbreak in a hematology-oncology unit associated with contaminated surface cleaning equipment. J Hosp Infect 2002; 52:93-98. 11. Becks VE, Lorenzoni NM. Pseudomonas aeruginosa outbreak in a neonatal intensive care unit: a possible link to contaminated hand lotion. Am J Infect Control 1995; 23:396-398.

outbreak of p. aeruginosa infection linked to icu design

33

12. Iversen BG, Jacobsen T, Hanne-Merete E, et al. An outbreak of Pseudomonas aeruginosa infection caused by contaminated mouth swabs. Clin Infect Dis 2007; 44:794-801. 13. Cobben NAM, Drent M, Jonkers M, Wouters EFM, Vaneechoutte M, Stobberingh EE. Outbreak of severe Pseudomonas aeruginosa respiratory infections due to contaminated nebulizers. J Hosp Infect 1996; 33:63-70. 14. Mount Sinai Hospital/Toronto Medical Laboratories Online Microbiology Lab Manual. Available at: http://microbiology.mtsinai.on.ca/manual/default.asp. Accessed March 28, 2008. 15. Gales AC, Reis AO, Jones RN. Contemporary assessment of antimicrobial susceptibility testing methods for polymyxin B and colistin: review of available interpretative criteria and quality control guidelines. J Clin Microbiol 2001; 39:183-190. 16. Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-eld gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995; 33:2233-2239. 17. 2006 Guidelines for Design and Construction of Health Care Facilities. 3.4.2 Critical Care Units (General). Washington, DC: American Institute of Architects; 2006:50. 18. Clark JA. Plumbing Engineer: plumbing xture selections for health care facilities. Available at: http://www.plumbingengineer.com/oct_06/xtures.php. Accessed March 28, 2008. 19. Levin MH, Olson B, Nathan C, Kabine SA, Weinstein RA. Pseudomonas in the sinks of an intensive care unit: relation to patients. J Clin Pathol 1984; 37:424-427. 20. Muscarella LF. Contribution of tap water and environmental surfaces to

21.

22.

23.

24.

25.

26.

27.

28.

nosocomial transmission of antibiotic-resistant Pseudomonas aeruginosa. Infect Control Hosp Epidemiol 2004; 25:342-345. Pitten FA, Panzig B, Schroder G, Tietze K, Kramer A. Transmission of a multiresistant Pseudomonas aeruginosa at a German university hospital. J Hosp Infect 2001; 47:125-130. Reuter S, Sigge A, Wiedeck H, Trautmann M. Analysis of transmission pathways of Pseudomonas aeruginosa between patients and tap water outlets. Crit Care Med 2002; 30:2222-2228. Trautmann M, Michalsky T, Wiedeck H, Radosavijevic V, Ruhnke M. Tap water colonization with Pseudomonas aeruginosa in a surgical intensive care unit and relation to Pseudomonas infections of ICU patients. Infect Control Hosp Epidemiol 2001; 22:49-52. Bonten MJM, Weinstein RA. Transmission pathways of Pseudomonas aeruginosa in intensive care units: dont go near the water. Crit Care Med 2002; 30:2384-2385. Berrouanne YF, NcNutt LA, Buschelman BJ, et al. Outbreak of severe Pseudomonas aeruginosa infections caused by contaminated drain in a whirlpool bathtub. Clin Infect Dis 2000; 31:1331-1337. Presteri E, Suchomel M, Eder M, et al. Effects of alcohols, povidoneiodine, and hydrogen peroxide on biolms of Staphylococcus epidermidis. J Antimicrob Chemother 2007; 60:417-420. Buckingham-Meyer K, Goeres DM, Hamilton MA. Comparative evaluation of biolm disinfectant efcacy tests. J Microbiol Methods 2007; 70: 236-244. Pitt TL. Epidemiological typing of Pseudomonas aeruginosa. Eur J Clin Microbiol Infect Dis 1988; 7:238-247.